Retroviral integration and the XMRV provirus

XMRVA strong argument that the novel human retrovirus XMRV is not a laboratory contaminant is the finding that viral DNA is integrated in chromosomal DNA of prostate tumors. Why does this result constitute such strong proof of viral infection?

Establishment of an integrated copy of the viral genome – the provirus – is a critical step in the life cycle of retroviruses. Proviral DNA is transcribed by cellular RNA polymerase II to produce the viral RNA genome and the mRNAs required to complete the replication cycle. Without proviral DNA, retroviral replication cannot proceed.

To produce proviral DNA, the retroviral RNA genome is converted to a double-stranded DNA by the viral enzyme reverse transcriptase. This step occurs in the cytoplasm. Specific and efficient insertion of the viral DNA into the host cell DNA is catalyzed by a viral enzyme called integrase. This enzyme recognizes and nicks the two ends of viral DNA, and the new 3′-ends are then joined covalently to the host DNA at staggered nicks made by integrase.

The image below shows some of the characteristic features of retroviral integration. A the top is the unintegrated linear DNA of avian sarcoma/leukosis virus produced by reverse transcription. Upon completion of integration, two base pairs (AA€¢TT) are lost from both termini, and a 6-bp target site in host DNA (pink) is duplicated on either side of the proviral DNA. This target site varies in length from 4 to 6 bp among different retroviruses. The proviral DNA (middle) ends with the conserved 5€²-T G€¦C A-3€² sequence. The provirus serves as a template for the production of the viral RNA genome (bottom).

To identify XMRV proviral DNA, genomic DNA was isolated from prostate tumors, and DNA was amplified using a primer that annealed in the viral env gene, near the right-hand LTR. Nucleotide sequence analyses of amplified DNAs from 14 9 different patients showed the expected viral CA sequence followed by human DNA. However, the other cardinal sign of retroviral integration – duplication of host DNA sequences flanking the integration site – could not be confirmed, because only the right-hand integration site was studied.

The isolation of the entire proviral DNA, including both flanking integration sites, from patients with prostate cancer or chronic fatigue syndrome would be additional evidence that XMRV is a virus that infects humans.

Kim, S., Kim, N., Dong, B., Boren, D., Lee, S., Das Gupta, J., Gaughan, C., Klein, E., Lee, C., Silverman, R., & Chow, S. (2008). Integration Site Preference of Xenotropic Murine Leukemia Virus-Related Virus, a New Human Retrovirus Associated with Prostate Cancer Journal of Virology, 82 (20), 9964-9977 DOI: 10.1128/JVI.01299-08

64 thoughts on “Retroviral integration and the XMRV provirus”

  1. Pingback: Tweets that mention Retroviral integration and the XMRV provirus --

  2. A thought-provoking post, as usual. One minor correction: they actually looked at samples from 9 patients (8 of them QQ mutants) and found 14 apparent integration sites among them. It looks like none of the patients returned a negative result in this assay.

    Besides looking at the other end to check for duplication, I’d also want to see some blinded samples from healthy controls. That’s especially important in this case, as they’re doing a preliminary linear amplification on the patient samples before doing the junction PCR. With that much amplification, and with the first step open-ended, I think there’s some room for PCR artifacts.

    Nonetheless, it’s a cool study, and I hope they’re following up on it.

  3. You state “The isolation of the entire proviral DNA, including both flanking integration sites, from patients with prostate cancer or chronic fatigue syndrome would be additional evidence that XMRV is a virus that infects humans.”

    If isolation of the entire proviral DNA with flanking sites indicating human DNA is present would only provide ‘additional evidence ‘ that XMRV is a human infectious virus – what would clinch the case for you? Why is that not sufficient to prove XMRV is a human infectious virus?

  4. Good stuff Dr. R. What can we expect from you with XMRV in the future?

    Are you collaborating with anyone? Just don’t collaborate with the pyscho brits unless you want to talk and go for a walk!

  5. Dr. Dove,

    A few questions to help a non-virologist understand your post, if possible. If these would take too much time to answer, it would be just fine if you could cite something publicly available that might help me understand the terms.

    What are QQ mutants and why is that interesting here?

    What does it mean that the researchers found “14 apparent integration sites” among the 9 patients?



  6. Dr. Rancaniello,

    Thank you so much for this explanatory blog post and especially the graphics. I had never planned to take up virology as a hobby, especially this late in life, so I very much appreciate your blogs and this explanation, in particular.


  7. Is it not also interesting that XMRV has been found integrated into NFATc3 and CREB5?
    And what is the significance of XMRV being shown to be integrated on one side? It cannot possibly be only on one side.

    I also look forward to tissue biopsies in ME/CFS patients.

  8. I don’t work on XMRV or any retrovirus – we work on other viruses. I
    write and podcast about XMRV and other viruses as a gratis public

  9. I’m taking the liberty of answering your questions, as I’m not sure if
    Alan will do it. QQ refers to an amino acid change in the protein
    ribonuclease L (RNAseL). This enzyme plays an important role in innate
    antiviral responses – the very early response to infection. Mutations
    in RNAseL, specifically at position 462 (R462Q, a change from arginine
    to glutamine) have been suggested to be susceptibility factors for
    prostate cancer. In the study described here, 8 of the 9 patients were
    homozygous for the Q change; e.g. they have a R to Q change in both
    alleles of the gene encoding RNAseL. The 14 apparent integration sites
    means that some of the patients had more than one integrated copy of

  10. XMRV integration near genes for NFATc3 and CREB5 could be interesting
    – they are involved in regulation of mRNA synthesis which could be
    perturbed by integration. But these integration sites were both found
    in one patient. The authors only looked for one of the XMRV
    integration sites, near the env gene – the other site should be

  11. Dr. Rancaniello,

    Now that viral DNA has been shown to be integrated in chromosomal DNA of prostate tumor (Eric Klein of Cleveland Clinic shown XMRV integration in cancerous prostate tissue) and the WPI has shown XMRV antibodies, do you now have an opinion about XMRV in light of the 4 contamination papers, other negative XMRV papers and positive XMRV papers?

  12. Summary:

    1) A strong argument that the novel human retrovirus XMRV is not a laboratory contaminant is the the finding that viral DNA is INTEGRATED in chromosomal DNA of prostate tumors(Klein, Silverman, et al).

    2) The isolation of the entire proviral DNA, including both flanking integration sites, from patients with prostate cancer or chronic fatigue syndrome would be ADDITIONAL evidence that XMRV is a virus that infects humans.

  13. Proviral DNA has been isolated from prostate tumors by one research
    group – Silverman. It must be confirmed by others. Such proviral DNA
    has not been isolated from clinical specimens obtained from patients
    with ME/CFS.

  14. Sarahphilips231

    So if the rhesus macaques and the shrewmouse get sick, thats it then finally people will believe that people with CFS are ill and its not a mental illness!?

  15. I have read your post and you have laid out some of the facts here nicely. Thank you! However, I am not clear about your professional opinion about XMRV thus far, if you have one at all with the limited and conflicting XMRV/prostate cancer/ME-CFS studies. I understand the rational need for replicating the proviral dna finding in prostate tumors by silverman as well as finding it in ME/CFS specimens. So it may be difficult to have an opinion about XMRV right now, yet many such as Dr Rein and Dr Mikovitz already hold strongly opposing opinions for example. Pardon me for asking.

  16. Complex retroviruses such as HIV-1 and HTLV-1 cause mitochondrial
    permeability, swelling and fragmentation during infection. I don’t
    know of any reports that such activities are observed in cells
    infected with simple retroviruses like the MLVs. These mitochondrial
    changes are related to induction of programmed cell death (apoptosis).
    I’m not aware of any reports that XMRV induces apoptosis in infected
    cells. Based on current data, it seems unlikely that XMRV would cause
    mitochondrial dysfunction. But it is well known that antiretroviral
    drugs against HIV cause mitochondrial toxicity.

  17. Perhaps I can help Dr R – lancelot is asking do you believe Alan Rein, Greg Towers, Simon Wessely and Dr Stoye who have already made up heir mind that there should be little if any more funding for XMRV, that XMRV is highly unlikely to be associated with ME/CFS.

    Or is it your professional opinion that there is now strong prima facie evidence for a link between XMRV and ME/CFS, that it is now looking unlikely that it is a contaminant, that therefore funding must be increased to put a stop to the terrible suffering of patients.

    Or are you sitting on the fence for the moment? In which case surely there should be more funding to determine the matter given the gravity of the what is at stake for patients?

  18. Based on the published data, I believe that XMRV, and possibly related
    MLVs, can infect humans. Whether XMRV infection causes human disease
    has not yet been proven. I do believe that additional research is
    needed to identify reliable assays to detect XMRV; to identify sites
    of replication in humans; to study viral replication in animal models;
    and to establish whether infection causes disease. There should be
    more funding of XMRV research to answer these questions.

  19. Thank you. So if CFS is correlated with mitochondrial dysfunction, then the treatment of CFS patients with antivirals may give mixed results, if the antivirals themselves cause mitochondrial toxicity?

  20. Thank you very much for your post Dr R, if I may ask one more question, how is causation best determined?

    Marshall infected and treated himself three times of H. pylori to prove it caused ulcers. Given the disbelief with respect to these viruses, are treatment trials the best way to prove causation with both symptomatic improvement and viral loads as objective evidence? Would more NRTI’s and Protease inhibitors be needed to have a highly active therapy as clear cut evidence of improvement would be needed given the disbelief.

    Also do you know of any drugs that were developed for HIV, yet failed to have significant effects on that virus, that may be usefl for XMRV? Is there anything in the bottom draw as it were?

  21. Do you think that existing anti-retroviral drugs would be effective against XMRV, if it were the cause of disease? If you were a clinician and had a CFS patient who tested positive for an MLV-related virus, would you prescribe them?

  22. These are great questions. Causation is best established in multiple
    ways. One is to do a case-control study with large numbers (thousands)
    of people in diverse geographic locations and determine if the virus
    is consistently associated with disease. This requires reliable assays
    for the virus which are not prone to contamination. A treatment trial
    would be another possibility, but if you want to use viral load as a
    measure of effectiveness, there has to be a good assay for that. The
    caveat here is that even if the virus caused the disease, the
    antivirals might not alleviate symptoms, for a variety of reasons. It
    would also be useful to know in what tissues the virus replicates in,
    to better understand how disease results. This requires sampling a
    variety of human tissues – not always possible unless done at autopsy.
    Animal models for infection can also play a big role in establishing
    causation, but I don’t know how one would establish ME/CFS in an
    animal model, given the importance of communication with the patient
    in diagnosis. Perhaps one could identify key sites of replication in
    animals, and biomarkers for disease, which could then be tested in
    humans. Just some of my ideas – a lot of speculation but all are
    possible. The idea of self-administering the infectious agent probably
    isn’t a good idea – you could not get approval to do that from your
    Institutional Review Board and if you tried to publish the findings
    you would be in big trouble.

  23. And to answer your last question – when the pharmaceutical companies
    screened for anti-HIV drugs there were thousands that failed but are
    still part of the chemical collections. These can be re-examined by
    screening using XMRV.

  24. We know that some anti-retroviral drugs will inhibit XMRV replication
    in cell culture. That does not mean that these drugs will alleviate
    disease (assuming, as you say, that XMRV causes disease). It depends
    on how the virus actually causes disease. For HIV, viral replication
    kills T cells, causing immunodeficiency and eventually death from
    opportunistic infections. We don’t know if XMRV destroys certain
    cells, leading to ME/CFS – we don’t even know where the virus
    replicates. One possibility is that the viral DNA integrates next to a
    cellular gene, activates it, and that causes disease. In that case, in
    a patient with disease, treating with an antiretroviral will not
    affect disease because viral replication is not the cause. I’m not a
    clinician, so I’m not qualified to speculate about treatment. Perhaps
    that question should be directed at infectious disease physicians in
    medical centers where research is done. Perhaps I’ll ask some of my
    colleagues that question.

  25. I asked six of my colleagues here who are MDs in the Infectious Disease division whether they would treat a MLV-positive, CFS patient with antivirals. Four said no, one said I don’t know, and the sixth has not answered.

  26. That answer from your colleagues is pretty reflective of the state of health care patients with ME/CFS get, Vincent. 9 times out of 10, you’re being shown the door, and they still collect the money. There are just a handful of specialists for ME/CFS in the whole world. After all, this disease is not attached to any speciality and hasn’t been recognized as of yet as an infectious disease.

  27. I would like to quote Dr Leonard Jason who commented on how patients with ME/CFS were treated in 2010: ” It’s fascinating if you think about it that we get hundreds of thousands of people who have an illness where they are really sensitive to stress and then we put them in a situation where they’re deprived of all types of resources have incredible needs and they’re incredibly distressed… we wouldn’t do that to laboratory mice. “

  28. I’ve been sent to two infectious disease doctors, and not only do they not believe ME/CFS is an infectious disease, they have also been at great pains to explain to me how it *couldn’t possibly* be an infectious disease. There don’t seem to be many infectious disease doctors out there who are even aware of new research in the field – and by “new” I mean from the last 15 years or so. One told me I had a form of anxiety disorder and needed antidepressants; the other said CFS was probably a dozen different diseases and was “too complicated” to have an infectious cause.

    I don’t think infectious disease doctors are obligated to start treating CFS patients with anti-retroviral drugs based on the current state of the science, and I wouldn’t want such treatment right now myself (unless it were part of a properly designed clinical trial.) But ID specialists’ ignorance of the fact that there *is* new science and legitimate debate going on kinda offends me.

  29. Yes Urbantravels, that is exactly what I have experienced. My UK based I.D. doctor clearly hasn’t taken the time to read the information I have supplied to her on XMRV. She therefore remains largely ignorant of the detailed science / facts about XMRV and consequently dissmissive of the association with ME/cfs. Unforgiveable.

  30. Since ME/CFIDS patients are known to have a weakness in the RNAseL pathway, could the finding of mutations at position 462 be considered a possible entry site for the XMRV family of viruses?

  31. That’s correct, in that ID docs would not necessarily be treating
    ME/CFS. But the ID docs I asked here are all involved in clinical
    research, all treat patients, and are all following the XMRV/ME/CFS
    story. Three of the individuals stated as their reason that there is
    no evidence of benefit for the antiretrovirals. If I can find an MD
    here with more experience in treating ME/CFS I’ll ask them what they
    would do.

  32. Great answers too, if I may say, Dr R! So it basically all boils down to a good reliable assay. Given that disease definition seems crucial to the question of causation too, do you know if there is a consensus to moving towards more specific (perhaps less sensitive) case definitions such as the Canadian Criteria?

    Or is there a consensus to predominantly do research on the severely affected where the symptoms are so severe that it cannot be mistaken for other conditions? I read the Lo/Alter paper where they found a much higher percent of positives(96%) in rigorously defined patients.

  33. I have been on AZT, TDF, RAL for 10 months with little or no improvement in my CFS. Those whom I have asked say that the disease is being caused by integrated provirus. Do these reservoirs decay over time on their own (if they are not being replenished)?

  34. The HIV virus was isolated from patients lymphadenopathy – the
    earliest patients were dying from opportunistic infections. A blood
    test was developed to detect the virus, then it was used to study
    hundreds, then thousands of patients with the disease – eventually
    called AIDS. Virus was consistently found in these patients, and
    chains of transmission were established. Furthermore there were
    transfusion acquired cases, and hospital acquired cases (eg needle
    stick with contaminated needle) that further proved the case. Finally
    when antiretrovirals were used they were shown to reverse the disease.
    That’s a simplified version of the history.

  35. I have had FM/CFS for 6years. Diagnosed by a series of MD’s. Recently I have began to feel a little better. The recipe was: Avoid all Antidepressants such as Effexor (SNRI’s) as they react strongly with pain killers such as Tramacet to produce Serotonin Syndrome (crazy making ugly). Back off or eliminate sleeping pills. Back off on Tylenol. Use Ibuprofen instead. Increase Vitamin C usage to more than 3000 mgs/day. Reduce stress. Do I believe that FM/CGS is anything other than a microbial problem? No. does stress aggrevate the symptoms? Yes. Can I convince my doctor that the disease is not the result of stress? No.

  36. Pingback: Authenticity of XMRV integration sites

Comments are closed.

Scroll to Top